Create README.md

README.md

@@ -0,0 +1,92 @@
+---
+model-index:
+- name: llama-3.1-tulu-2-70b-uf-mean-rm
+  results: []
+datasets:
+- allenai/tulu-2.5-preference-data
+- allenai/tulu-v2-sft-mixture
+language:
+- en
+base_model: allenai/llama-3.1-tulu-2-70b
+license: apache-2.0
+---
+<center>
+<img src="https://huggingface.co/datasets/allenai/blog-images/resolve/main/tulu-2.5/tulu_25_banner.png" alt="Tulu 2.5 banner image" width="800px"/>
+</center>
+
+# Model Card for Llama 3.1 Tulu V2 70B RM - UltraFeedback
+
+Tulu is a series of language models that are trained to act as helpful assistants.
+This is a 70B reward model used for PPO training trained on the UltraFeedback dataset.
+
+For more details, read the paper:
+[Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback](https://arxiv.org/abs/2406.09279).
+
+Note this model is finetuned from Llama 3.1, released under the Meta Llama 3.1 community license, included here under `llama_3_license.txt`.
+
+
+## Performance
+
+We evaluate the model on [RewardBench](https://github.com/allenai/reward-bench):
+
+| Model            | Score | Chat  | Chat Hard | Safety | Reasoning |
+|------------------|-------|-------|-----------|--------|-----------|
+| [Llama 3.1 Tulu 2 8b UF RM](https://huggingface.co/allenai/llama-3.1-tulu-2-8b-uf-mean-rm)  | 73.3  | 98.0  |    59.6   |  60.6  |    74.7   |
+| **[Llama 3.1 Tulu 2 70b UF RM](https://huggingface.co/allenai/llama-3.1-tulu-2-70b-uf-mean-rm) (this model)** |     | |  |  |  |
+
+
+## Model description
+
+- **Model type:** A reward model trained on UltraFeedback, designed to be used in RLHF training.
+- **Language(s) (NLP):** English
+- **License:** Apache 2.0.
+- **Finetuned from model:** [allenai/llama-3.1-tulu-2-70b](https://huggingface.co/allenai/llama-3.1-tulu-2-70b)
+
+### Model Sources
+
+- **Repository:** https://github.com/allenai/open-instruct
+- **Dataset:** Data used to train this model can be found [here](https://huggingface.co/datasets/allenai/tulu-2.5-preference-data) - specifically the `ultrafeedback_mean_aspects` split.
+
+
+## Input Format
+
+The model is trained to use the following format (note the newlines):
+```
+<|user|>
+Your message here!
+<|assistant|>
+```
+
+For best results, format all inputs in this manner. **Make sure to include a newline after `<|assistant|>`, this can affect generation quality quite a bit.**
+We have included a [chat template](https://huggingface.co/docs/transformers/main/en/chat_templating) in the tokenizer implementing this template.
+
+## Intended uses & limitations
+
+The model was initially fine-tuned on a filtered and preprocessed of the [Tulu V2 mix dataset](https://huggingface.co/datasets/allenai/tulu-v2-sft-mixture), which contains a diverse range of human created instructions and synthetic dialogues generated primarily by other LLMs. 
+We then further trained the model with a [Jax RM trainer](https://github.com/hamishivi/EasyLM/blob/main/EasyLM/models/llama/llama_train_rm.py) built on [EasyLM](https://github.com/young-geng/EasyLM) on the dataset mentioned above.
+This model is meant as a research artefact.
+
+### Training hyperparameters
+
+The following hyperparameters were used during RM training:
+- learning_rate: 5e-06
+- total_train_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear cooldown to 0.
+- lr_scheduler_warmup_ratio: 0.03
+- num_epochs: 1.0
+
+## Citation
+
+If you find Tulu 2.5 is useful in your work, please cite it with:
+
+```
+@misc{ivison2024unpacking,
+      title={{Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback}}, 
+      author={{Hamish Ivison and Yizhong Wang and Jiacheng Liu and Ellen Wu and Valentina Pyatkin and Nathan Lambert and Yejin Choi and Noah A. Smith and Hannaneh Hajishirzi}}
+      year={2024},
+      eprint={2406.09279},
+      archivePrefix={arXiv},
+      primaryClass={cs.CL}
+}
+```